Loading ...
Sorry, an error occurred while loading the content.

GEOSTATS: Geostats vs. Spatial Stats?

Expand Messages
  • Justin_Quirouette@pch.gc.ca
    Hello, I seem to recall coming across a differenciation between geostatistics and spatial statistics . Could someone provide me with links defining and
    Message 1 of 6 , Sep 28, 1998
    • 0 Attachment
      Hello,


      I seem to recall coming across a differenciation between
      "geostatistics" and "spatial statistics". Could someone provide me
      with links defining and describing the two?


      Thank You in advance. Will SUM.
      Justin Q.



      Justin Quirouette
      GIS Specialist
      Georgian Bay Islands National Park
      Honey Harbour, Ontario
      CANADA


      --
      *To post a message to the list, send it to ai-geostats@....
      *As a general service to list users, please remember to post a summary
      of any useful responses to your questions.
      *To unsubscribe, send email to majordomo@... with no subject and
      "unsubscribe ai-geostats" in the message body.
      DO NOT SEND Subscribe/Unsubscribe requests to the list!
    • Gregoire Dubois
      ... Many authors who have a background in geology would tell that geostatistics is based on the theory of Regionalised variables developed by Matheron while
      Message 2 of 6 , Sep 29, 1998
      • 0 Attachment
        Justin_Quirouette@... wrote:
        >
        > Hello,
        >
        > I seem to recall coming across a differenciation between
        > "geostatistics" and "spatial statistics". Could someone provide me
        > with links defining and describing the two?


        Many authors who have a background in geology would tell that
        geostatistics is based on the theory of Regionalised variables
        developed by Matheron while other spatial statistics are not.
        Because it has been initially mainly applied to spatial estimations
        in the mining industry, it has been called Geostatistics.
        Cressie (Statistics for spatial data, page 29) gives however a more
        larger description of the origin of the word geostatistics, showing
        that the prefix geo was initially refering to geographical data
        (spatial data in the broad sense), then was put in a geological context
        with Matheron. Cressie concludes that for him "geostatistics has thrown
        off its earthly shackles and has taken on a more universal role, one that
        is concerned with statistical theory and application for processes with
        continuous spatial index".

        In general, we refer to geostatistics when the statistics are applied
        to a spatial phenomenum that is continuous over space (geostatistical
        data are for example rainfall data, mineral concentrations, ...), and
        to spatial statistics when statistics are applied to point patterns
        (the location itself is the variable of interest, this means clustering
        analysis, spatial randomness analysis,....) and to lattice data (regularly
        spaced data, like images provided by remote sensing techniques).

        Hope this clarifies a bit the ambiguity.

        Best regards

        Gregoire



        --

        Gregoire Dubois Tel. 39-(0)332-78.99.44
        Joint Research Centre Fax. 39-(0)332-78.54.66
        Environment Inst. TP 321 Email: gregoire.dubois@...
        I-21020 Ispra (Va), ITALY URL: http://curie.ei.jrc.it/
        --
        *To post a message to the list, send it to ai-geostats@....
        *As a general service to list users, please remember to post a summary
        of any useful responses to your questions.
        *To unsubscribe, send email to majordomo@... with no subject and
        "unsubscribe ai-geostats" in the message body.
        DO NOT SEND Subscribe/Unsubscribe requests to the list!
      • Syed Abdul Rahman
        A popular taxonomy is that of Noel Cressie s (in _Statistics for Spatial Data_, Wiley Interscience) in which he gives three separate treatments for: (a)
        Message 3 of 6 , Sep 29, 1998
        • 0 Attachment
          A popular taxonomy is that of Noel Cressie's
          (in _Statistics for Spatial Data_, Wiley Interscience)
          in which he gives three separate treatments for:
          (a) geostatistical data (data with continuous
          spatial index); (b) lattice data (data with a countable
          collection of spatial sites, frequently regular); and
          (c) point pattern data (data with random spatial
          index). All of this under the umbrella of spatial data
          analysis.

          The differentiation is usually more academic than useful;
          such methods can (and do) overlap; e.g. the use of
          point pattern and lattice techniques on "geostatistical
          data."

          Regards,

          Syed

          Justin_Quirouette@... wrote:
          >
          > Hello,
          >
          > I seem to recall coming across a differenciation between
          > "geostatistics" and "spatial statistics". Could someone provide me
          > with links defining and describing the two?
          >
          > Thank You in advance. Will SUM.
          > Justin Q.
          >
          > Justin Quirouette
          > GIS Specialist
          > Georgian Bay Islands National Park
          > Honey Harbour, Ontario
          > CANADA
          >
          > --
          > *To post a message to the list, send it to ai-geostats@....
          > *As a general service to list users, please remember to post a summary
          > of any useful responses to your questions.
          > *To unsubscribe, send email to majordomo@... with no subject and
          > "unsubscribe ai-geostats" in the message body.
          > DO NOT SEND Subscribe/Unsubscribe requests to the list!


          --
          *To post a message to the list, send it to ai-geostats@....
          *As a general service to list users, please remember to post a summary
          of any useful responses to your questions.
          *To unsubscribe, send email to majordomo@... with no subject and
          "unsubscribe ai-geostats" in the message body.
          DO NOT SEND Subscribe/Unsubscribe requests to the list!
        • Jennifer Dungan
          ... In my opinion, Cressie s taxonomy is quite confusing. I think a more useful taxonomy is based on *models* applied to data, not on the data themselves
          Message 4 of 6 , Sep 30, 1998
          • 0 Attachment
            On Tue, 29 Sep 1998, Syed Abdul Rahman wrote:

            > A popular taxonomy is that of Noel Cressie's
            > (in _Statistics for Spatial Data_, Wiley Interscience)
            > in which he gives three separate treatments for:
            > (a) geostatistical data (data with continuous
            > spatial index); (b) lattice data (data with a countable
            > collection of spatial sites, frequently regular); and
            > (c) point pattern data (data with random spatial
            > index). All of this under the umbrella of spatial data
            > analysis.

            > > I seem to recall coming across a differenciation between
            > > "geostatistics" and "spatial statistics". Could someone provide me
            > > with links defining and describing the two?


            In my opinion, Cressie's taxonomy is quite confusing. I think
            a more useful taxonomy is based on *models* applied to data, not
            on the data themselves (data can't be inherently "geostatistical,"
            though they can be spatial).

            To answer Justin's question, I would say geostatistics is a subset of
            spatial statistics. The latter consists of any models that explicitly
            recognize spatial location in their construction, a very broad definition
            that includes a lot of different types of models. Geostatistics is
            founded on regionalized variable theory, a particular model (though with
            many manifestations and flavors).

            There is a "deterministic" interpretation of geostatistics that does not
            make reference to regionalized variable theory and random functions (i.e.
            Journel (1985) 17:1, and Armstrong (1985) Math Geol 17:855). This
            interpretation may expand the definition of geostatistics to refer
            to a larger subset of spatial statistical models.

            Jennifer Dungan | MS 242-4
            CSU - Monterey Bay | NASA Ames Research Center
            Tel: 650-604-3618 FAX: 650-604-4680 | Moffett Field, CA 94035-1000
            email: jdungan@... | USA
            URL: http://geo.arc.nasa.gov |

            --
            *To post a message to the list, send it to ai-geostats@....
            *As a general service to list users, please remember to post a summary
            of any useful responses to your questions.
            *To unsubscribe, send email to majordomo@... with no subject and
            "unsubscribe ai-geostats" in the message body.
            DO NOT SEND Subscribe/Unsubscribe requests to the list!
          • Duane F. Marble
            ... Thus, multi-regional mathematical programming models (clearly spatial) would also be statistical in this definition. My view has always been that spatial
            Message 5 of 6 , Sep 30, 1998
            • 0 Attachment
              You note that:

              >To answer Justin's question, I would say geostatistics is a subset of
              >spatial statistics. The latter consists of any models that explicitly
              >recognize spatial location in their construction, a very broad definition
              >that includes a lot of different types of models.

              Thus, multi-regional mathematical programming models (clearly spatial)
              would also be "statistical" in this definition. My view has always
              been that spatial statistics was one area within the broader arena
              of spatial analysis that covers both statistical and non-statistical
              spatial models. Or are we working with a new, global definition of
              "statistics?"

              ***********************************************************************
              Professor Emeritus Duane F. Marble
              The Center for Mapping and
              the Department of Geography Email: marble.1@...
              The Ohio State University
              Columbus, Ohio 43212-1154

              Phone: (614) 292-3409 (CFM) or 292-2250 (Geography)
              Fax: (614) 292-8062 (CFM) or 292-6213 (Geography)

              --
              *To post a message to the list, send it to ai-geostats@....
              *As a general service to list users, please remember to post a summary
              of any useful responses to your questions.
              *To unsubscribe, send email to majordomo@... with no subject and
              "unsubscribe ai-geostats" in the message body.
              DO NOT SEND Subscribe/Unsubscribe requests to the list!
            • Jennifer Dungan
              ... No. I should have qualified the above with statistical. As in, The latter consists of any statistical models that explicitly recognize spatial
              Message 6 of 6 , Sep 30, 1998
              • 0 Attachment
                On Wed, 30 Sep 1998, Duane F. Marble wrote:

                > You note that:
                >
                > >To answer Justin's question, I would say geostatistics is a subset of
                > >spatial statistics. The latter consists of any models that explicitly
                > >recognize spatial location in their construction, a very broad definition
                > >that includes a lot of different types of models.
                >
                > Thus, multi-regional mathematical programming models (clearly spatial)
                > would also be "statistical" in this definition. My view has always
                > been that spatial statistics was one area within the broader arena
                > of spatial analysis that covers both statistical and non-statistical
                > spatial models. Or are we working with a new, global definition of
                > "statistics?"

                No. I should have qualified the above with "statistical." As in,
                "The latter consists of any statistical models that explicitly recognize
                spatial location..."

                -Jennifer

                Jennifer Dungan | MS 242-4
                CSU - Monterey Bay | NASA Ames Research Center
                Tel: 650-604-3618 FAX: 650-604-4680 | Moffett Field, CA 94035-1000
                email: jdungan@... | USA
                URL: http://geo.arc.nasa.gov |



                --
                *To post a message to the list, send it to ai-geostats@....
                *As a general service to list users, please remember to post a summary
                of any useful responses to your questions.
                *To unsubscribe, send email to majordomo@... with no subject and
                "unsubscribe ai-geostats" in the message body.
                DO NOT SEND Subscribe/Unsubscribe requests to the list!
              Your message has been successfully submitted and would be delivered to recipients shortly.